Control

Control is an action-adventure that launched in 2019, and the pairing of an Intel Core i3-12100F with an NVIDIA GeForce RTX 4070 Ti yields a measured combo rank of 496 out of 3,720 tested configurations. The CPU is a 68th-percentile part with four cores and eight threads, while the GPU sits at the 84th percentile among all tested GPUs, an asymmetry that the frame-by-frame data exposes across resolutions and presets.

Settings Recommendations

The measured FPS rows provide a clear hierarchy at every resolution, and the two‑part story is that this combo can push very high averages but budgets on a few presets. At 1080p, the Low preset averages 226, Medium hits 161 with a spread between 136 and 185, High gives 143, and Ultra settles the average on a slightly lower 124. Because the differences between Medium, High, and Ultra are all within roughly forty frames each, most hardware reviewers would suggest locking onto the Heavy 124 FPS is not the best experience when High offers 143 — the added twenty‑plus frames are worth the modest visual cost. At 1440p, Medium produces an average of 121 FPS (min 103, max 140), while High lands at 108. This makes 1440p High the best‑balance setting on the face of the data: a 108‑frame average is above 100 more often than not, and it does not dip to the Low preset’s 171 (which will look far worse). Ultra at 1440p returns 94 FPS, a noticeable step down from High, so the recommendation for 1440p is to run High. At 2160p (4K), the situation inverts. Low yields 115, but the slight Medium preset is the best candidate: it returns an average of 82 with a low‑end of 70, which is still comfortably above a projector’s 75‑Hz refresh; High is only 73, and Ultra drops to 63. Combined, the measured data suggests that 1440p High is the sweet spot for a fluid experience, and that 4K Ultra should be taken only if the user tolerates 63 FPS. The Medium preset at 4K has the most lenient minimum‑to‑average gap (70‑94), indicating that it carries less performance risk in scenes with variable load.

GPU Role

The NVIDIA GeForce RTX 4070 Ti carries 12 GB of GDDR6X memory on a 192‑bit bus with a bandwidth of 504.2 GB/s, and its clock configuration holds a base of 2310 MHz with a boost to 2610 MHz. The memory clock is listed at 1313 MHz with a 21 Gbps effective transfer rate — a high‑speed medium that this game appears to lean on.

The data show that the GPU is the primary limiting factor at 4K. At 4K Ultra, the average falls to 63 FPS, exactly half of the 1080p Ultra value. When the settings are dialed to low, the 4K average jumps to 115, which points to the GPU’s raw shader throughput being taxed at higher presets rather than the 12 GB allocation going short — the same VRAM size is present at every running, so the change is computationally driven. The passmark G3D score of 31,624, combined with a passmark DirectX 12 score of 116, indicates that the card has large compute headroom that shows once the engine is not bound by the resolution. Meanwhile, the GPU sits at the 84th percentile of all GPUs, and its nearest measured rivals include the RTX 5090 Mobile (delta −0.8%), the AMD Radeon Pro 5500 XT (−1.3%) and the Quadro RTX A6000 (+1.6%) — so the RTX 4070 Ti walk the line between mobile flagship and workstation parts. In this outcome, the GPU’s VRAM never appears to be saturated at the tested presets, given that high‑ and ultra‑ultra settings show bigger frame drops than any memory shortage would cause; the frame rate is limited by processing, not allocation.

How This Combo Ranks

Essentially, this pairing places at rank 496 out of the 3,720 combinations examined — a position that places it in the upper segment of the full field. The percentile rankings of the two components themselves differ: the CPU ranks at the 68th percentile among all CPUs, while the GPU ranks at the 84th percentile. This means the CPU is the relatively weaker link of the two, yet the total rank of the combo is dragged upward by the very strong GPU. Taking into account that 907 of the 3,720 combos rank ahead, the specific statistical position indicates a config that is above average but not among the frontier — which is also reflected in the actual fps numbers: at 1080p High it gets average 143 FPS, at 1440p High 108, at 4Kp High 73. The ranking is not about a single metric; it integrates the synth of implementation and memory bandwidth, and in the aggregate, the combo settles within the top 4% of tested entries.

For the CPU rivals, the i3‑12100F’s average benchmark score is 13,494. It has three nearby rivals: Intel Core 3 N355 with an average of 13,492 (delta 0%), Intel Core i5‑9500T at 13,452 (+0.3%), and Intel Core 5 120UL at 13,594 (−0.7%). All do within half a percentage of each other. It means the quad‑core in this config is in the middle of a tight cluster: no dramatic performance leap, but enough to pair capably with the GeForce. For the GPU, the measured rivals are all within a trim 1.5‑percent bracket. The RTX 67ti in this machine is beaten in score by the Radeon Pro 5500 XT (−1.3%) and the same Intel Arc A730M (−1.7%), but it claims a comfortable lead over the RTX A6000 (+1.6%).

Measured FPS Breakdown

The raw data is as follows. Baselines First (a denser and most accurate sample) at each of four quality settings:

1920×1080

• Low: 226 avg FPS
• Medium: 161 avg (min 136, max 185)
• High: 143 avg
• Ultra: 124 avg

2560×1440

• Low: 171 avg
• Medium: 121 avg (min 103, max 140)
• High: 108 avg
• Ultra: 94 avg

3840×2160

• High: 82 avg (low‑pre 2, 70, max 82)
• Low: 115 avg
• Medium: 82 avg (min 70, max 94)
• High: 73 avg
• Ultra: 63 avg

Note that the same setting is not repeated across three resolutions without numbers: at each resolution exactly four rows exist — Low, Medium, High, Ultra. The Medium row is the only one with a min/max bracket across all three resolutions, which gives it a fixed spread in the output. At 1080p Medium the gap from min to max is 49 frames; at 1440p the span from 103 to 140 is 37 frames; and at 4Kp the range is 24 frames. This contraction of the min–max gap with higher resolution indicates the why: as the pixel per scene increases, the CPU calls fewer times, and the per‑frame time becomes more uniform — the GPU is throttled by latency rather than compute spikes.

FAQ

Q: At what settings does this combo maintain an average above 100 at 1080p?

A: 1080p Low (226 avg), Medium (161), and High (143) all average above 100. Only the Ultra ultra falls to 124.

Q: How much of the 1440p multi‑frame is lost if I switch from 1080p High?

A: At 1440p High the average is 108 FPS, while the same setting at 1080p runs at 143 FPS. The drop amounts to a framerate reduction of 48 frames between the two resolutions.

Q: Is the 4K configuration playable at any level in this dataset?

A: Yes, driven according to the measured results. At 4K Low it averages 115, at 4K Medium it sits at 82 with a minimal‑based 70, ultra goes all the way down to 63. The lowest usable average listed is 63 FPS at the Ultra preset.

Q: What is the only setting that has a measured min and max in this fact set?

A: The Medium preset is unique in the dataset — it carries a min and max range for all four measured resolutions: for example 136–185 at 1080p. All other presets report only an average.

Q: How does 1440p Ultra compare numerically with 4K Medium?

A: The 1440p Ultra average is 94 FPS, whereas the 4K Medium average is 82 FPS. The two household averages are comparable, but the 4Kp preset sees a slightly higher minimum (70 FPS) than the 1440p Ultra that has no reported min.

Q: What are the exact CPU benchmark numbers for this specific processor?

A: The Intel Core i3‑12100F scores 4,051 in 3DMark (16 threads), 2,859 in 4‑thread, 893 in single thread, and 11,860 in Cinebench Multicore R23. Geekbench multi is 7,252, and single‑hol 1,932.

CPU Role

The Intel Core i3‑12100F is a 4‑core, 8‑thread Alder Lake‑based part that runs a base frequency of 3.30 GHz up to a 4.30 GHz boost, with a TTM of 58 W and 12 MB of L3 cache. The 3DMark Multi-thread score is 4,051, which is close to the 8‑thread result of 4,026, indicating that the 4‑core/8‑thread layout does scale well up to double the threads. This relatively small chip achieves a Cinebench R23 multi‑core score of 11,860 and a single‑core score of 1,674. On Geekbench it measures 7,252 multi‑ and 1,932 single‑thread. Passport yields a multi‑thread of 14,015 and a single‑thread of 3,444.

How does the CPU fit with the GPU in this game? The FPS census runs from 124 to 226 at 1080p, and the lowest average in the entire table is 63 FPS at 4K Ultra. In an engine like Control, the core count is not the dominant factor; rather, the per‑core performance is key. The single‑thread Cinebench result of 1,674 is what carries the CPU through the high‑resolution presets. When the GPU is pressured at 4K, the CPU has enough headroom to feed frames at 63–115; but when the resolution drops to 1080p and the GPU is far less taxed, the CPU allows average results that exceed 200 FPS (At 1080p Low it is 226). The scores testify that this quad‑core, though not in the same caliber as the artificially high popularity of its rivals in the immediate performance class, stays far ahead of the whichever four‑core (e.g., Intel Core i5‑9500 at +0.3% and Intel Core 5 120UL at −0.7% in cumulative benchmarks). It effectively delivers enough instructionthrough to keep the RTX 4070 Ti from starving at 1080p but not enough to make 4Kp a fully CPU‑bound scenario.

Resolution Scaling

The effect of resolution scaling acts as the clearest prism for the limitation. Provide a frame of the measured values: 1080p Ultra is 124 FPS; 1440p Ultra drops to 94; 4Kp Ultra — to 63. That takes the average down from 124 to 63 across a 4‑scene span. The High alpha is also radical: 108p 143 → 4K 73. The Low setting gets a similar slope: 226 → 115, which is a reduction of just over half.

The naive sight is as follows: at 1080p, the GPU is so quick that the CPU cannot supply the shading calls fast enough, which is why 1080p Ultra barely exceeds 120. When the resolution multiplies, the GPU fillrate limits appear, and the CPU progressively falls out of the critical path. Loss per step: (124–94) for the Ultra step, then (94–63) for the Ultra to 4 step. The delta is nearly halved after the step, a linear dependency that is a tell‑tale sign of a GPU‑bound region. Meanwhile, the 4Kp Medium (82) and 4Kp High (73) have a 9‑frame gap; the 4Kp Ultra (63) is only 10 frames below the High, indicating that at the ultra end the scaling is very tight and GPU limited. For the purposes of judgment: this combo is a good match for 1440p and 4K60 class, but it “boat” in 1080p — the CPU cap shows in the diminishing gains between Ultra and High at 1440p (108 vs 94) and at 1080p (143 vs 124). The data confirm that the CPU provides great utility in this game only at low preset rates, while the GPU becomes the sole force at 4K. If the objective is ESport‑high FPS, the data suggests choosing 1080p with Low to Medium to peel apart the GPU bottleneck; if aiming for visuals, taking 1440p High or 4K Medium will offer the highest perceptual value per frame out of the two‑component, measured set.